DeepSeek V4 Flash API: pricing, base URL and quick setup

Specifications and pricing

USD per 1M tokens. Final billing depends on your API key group and metered usage.

Cheapest way to use DeepSeek V4 Flash

Quick setup

curl https://a2agent.me/v1/chat/completions \
  -H "content-type: application/json" \
  -H "authorization: Bearer sk-your-a2agent-key" \
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [{"role": "user", "content": "Reply with OK"}]
  }'

Using it from a coding client

In Cline and Roo Code, track total tokens across the session, since the cost adds up across requests. Watch the output rate rather than the input one if your prompts are short and the answers run long.

• Run it in Cline
• Run it in Roo Code

How it compares within its family

These comparisons use catalog context limits, tags and reference rates, not independent quality benchmarks. Rates can change. Check current pricing above and test your workload; a higher price or a larger context window does not establish better results.

V4 Flash costs less than V4 Pro here, and price is the measured difference in this reference data. Context is 1M on both. The reasoning tag is on both. Flash lists at $0.14 per million input tokens against $0.435 for V4 Pro, and $0.28 output against $0.87. At that spread you can run a task three times on Flash for the price of one Pro call, which changes what is worth trying: sampling several answers and picking, or re-running a failed parse, stops being expensive. Neither rate establishes answer quality. Use a reviewed task set before relying on either model. Record input and output tokens separately so differences in answer length are included in the comparison.

Other models on this platform

• DeepSeek V4 Pro1M
• DeepSeek V4.1 Flash1M

Questions